- update LangchainJS version

- add types & interfaces

- small ui bug fix

- update marketplace files

packages/components/nodes/agents/ConversationalAgent/ConversationalAgent.ts

@@ -1,4 +1,9 @@
 import { INode, INodeData, INodeParams } from '../../../src/Interface'
+import { initializeAgentExecutor, AgentExecutor } from 'langchain/agents'
+import { Tool } from 'langchain/tools'
+import { BaseChatModel } from 'langchain/chat_models/base'
+import { BaseChatMemory } from 'langchain/memory'
+import { getBaseClasses } from '../../../src/utils'
 
 class ConversationalAgent_Agents implements INode {
     label: string
@@ -17,6 +22,7 @@ class ConversationalAgent_Agents implements INode {
         this.category = 'Agents'
         this.icon = 'agent.svg'
         this.description = 'Conversational agent for a chat model. It will utilize chat specific prompts'
+        this.baseClasses = [this.type, ...getBaseClasses(AgentExecutor)]
         this.inputs = [
             {
                 label: 'Allowed Tools',
@@ -37,16 +43,10 @@ class ConversationalAgent_Agents implements INode {
         ]
     }
 
-    async getBaseClasses(): Promise<string[]> {
-        return ['AgentExecutor']
-    }
-
     async init(nodeData: INodeData): Promise<any> {
-        const { initializeAgentExecutor } = await import('langchain/agents')
-
-        const model = nodeData.inputs?.model
-        const tools = nodeData.inputs?.tools
-        const memory = nodeData.inputs?.memory
+        const model = nodeData.inputs?.model as BaseChatModel
+        const tools = nodeData.inputs?.tools as Tool[]
+        const memory = nodeData.inputs?.memory as BaseChatMemory
 
         const executor = await initializeAgentExecutor(tools, model, 'chat-conversational-react-description', true)
         executor.memory = memory
@@ -54,7 +54,7 @@ class ConversationalAgent_Agents implements INode {
     }
 
     async run(nodeData: INodeData, input: string): Promise<string> {
-        const executor = nodeData.instance
+        const executor = nodeData.instance as AgentExecutor
         const result = await executor.call({ input })
 
         return result?.output